Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

I-vector vector extraction method and apparatus, speaker recognition method and device, equipment and medium

A technology for speaker recognition and extraction methods, applied in the field of speech recognition, can solve the problems of increasing the complexity of i-vector vectors, and achieve the effect of reducing the complexity of recognition and improving the purity.

Active Publication Date: 2018-12-21
PING AN TECH (SHENZHEN) CO LTD
View PDF9 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are still many interference factors in the i-vector vector obtained by the existing i-vector modeling, which increases the complexity of using the i-vector vector for speaker recognition

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • I-vector vector extraction method and apparatus, speaker recognition method and device, equipment and medium
  • I-vector vector extraction method and apparatus, speaker recognition method and device, equipment and medium
  • I-vector vector extraction method and apparatus, speaker recognition method and device, equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0040] The i-vector vector extraction method provided by the embodiment of the present invention can be applied in such as figure 1 An application environment in which a computer device communicates with a recognition server over a network. Among them, computer equipment includes but is not limited to various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices. The identification server can be implement...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an i-vector vector extraction method and apparatus, a speaker recognition method and device, equipment and a medium. The i-vector vector extraction method comprises: acquiringtraining speech data of a speaker and extracting training speech features corresponding the training speech data; on the basis of a preset UBM model, training an overall changing subspace corresponding to the preset UBM model; projecting the training speech data on the overall changing subspace and acquiring a first i-vector vector; and projecting the first i-vector vector on the overall changingsubspace and acquiring a registered i-vector vector corresponding to the speaker. According to the method disclosed by the invention, the training speech feature data are projected twice, the dimension can be reduced, more noise features can be removed, and the purity of the speech feature of the speaker can be improved; and after dimension reduction, the computing space is reduced and the recognition efficiency of the speech recognition is also improved.

Description

technical field [0001] The invention relates to the field of speech recognition, in particular to an i-vector vector extraction method, speaker recognition method, device, equipment and medium. Background technique [0002] Speaker recognition, also known as voiceprint recognition, is a biometric authentication technology that uses specific speaker information contained in voice signals to identify the identity of the speaker. In recent years, the introduction of i-vector (identity-vector, identity authentication vector) modeling method based on vector analysis has significantly improved the performance of speaker recognition systems. In the vector analysis of the speaker's speech, the information of the speaker is usually included in the channel subspace. The i-vector space uses a low-dimensional total variable space to represent the speaker subspace and the channel subspace, and the speaker's voice is projected into the space through dimensionality reduction, and a fixed-...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/02G10L15/06G10L25/51
CPCG10L15/02G10L15/063G10L25/51G10L2015/0638
Inventor 涂宏
Owner PING AN TECH (SHENZHEN) CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products